Are two informants better than one? Parent-child agreement on the eating styles of children who are overweight.

AIM

It is currently unknown to what extent the view of a child with overweight on its' own eating behaviour converges with parental perception regarding this behaviour and how parent-child agreement is influenced by overweight status and age.

METHOD

Youngsters (N = 498; range 7-15 years; 37% boys) referred for weight treatment to an outpatient University centre filled in the Dutch Eating Behaviour Questionnaire-child version (DEBQ-child version; Van Strien & Braet, unpublished work), prior to treatment, while their parents reported on their child's eating behaviour by completing the DEBQ-parent version (Braet & Van Strien, 1997).

RESULTS

Parents scored significantly higher when reporting on the emotional eating and external eating behaviour of their child, while they scored lower for restrained eating (all p < 0.001). Comparisons between the subscales of the DEBQ-parent version and the DEBQ-child version revealed significant positive correlations of r = 0.45 for emotional eating, r = 0.35 for external eating and r = 0.36 for restrained eating (all p < 0.01); convergence is lowest for the age group younger than 10 (p < 0.05). Both versions of the DEBQ displayed low correlations with the degree of overweight of the child.

DISCUSSION

Parents and children displayed moderate to good agreement with regard to emotional eating, external eating and restrained eating. However when only one perspective can be assessed, possible biases must be taken into account. In that case, the use of appropriate age-specific norms is indicated.